CN106713926A - Compression storage method and device for video data - Google Patents
Compression storage method and device for video data Download PDFInfo
- Publication number
- CN106713926A CN106713926A CN201611237039.0A CN201611237039A CN106713926A CN 106713926 A CN106713926 A CN 106713926A CN 201611237039 A CN201611237039 A CN 201611237039A CN 106713926 A CN106713926 A CN 106713926A
- Authority
- CN
- China
- Prior art keywords
- data block
- compression
- data
- storage
- sign
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
- H04N19/42—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals characterised by implementation details or hardware specially adapted for video compression or decompression, e.g. dedicated software implementation
- H04N19/423—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals characterised by implementation details or hardware specially adapted for video compression or decompression, e.g. dedicated software implementation characterised by memory arrangements
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
- H04N19/10—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding
- H04N19/134—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the element, parameter or criterion affecting or controlling the adaptive coding
- H04N19/156—Availability of hardware or computational resources, e.g. encoding based on power-saving criteria
Landscapes
- Engineering & Computer Science (AREA)
- Multimedia (AREA)
- Signal Processing (AREA)
- Computing Systems (AREA)
- Theoretical Computer Science (AREA)
- Compression Or Coding Systems Of Tv Signals (AREA)
- Compression, Expansion, Code Conversion, And Decoders (AREA)
Abstract
The invention provides a compression storage method for video data. The method comprises the following steps: after compressing the video data, obtaining a data block header and a data block body; storing the data block header and the data block body in a predetermined storage space together; setting an indication identifier for indicating the corresponding data block compression size and storing the data block in a memory in a fixed bit number according to raster scan; and based on a bus bandwidth, determining a storage basic unit of the compressed data block, and storing the compressed data block according to the storage basic unit and the raster scan sequence. The invention further provides a storage device for storing the video data. Under the same performance requirement, the required cache space is 1/4 of the original cache space, and meanwhile, the number of cache paths is 1/4 of the original cache paths, and the number of the cache paths is greatly reduced, which directly results in the reduction of the hardware implementation logic level, and is favorable for back-end timing convergence and can run to a higher frequency and significantly improve the performance.
Description
Technical field
The invention belongs to video data compression processing technology field, more particularly to a kind of compression and storage method of video data
And device.
Background technology
Frame buffer compression (FBC, Frame buffer compression) technology is based on current bandwidth applications scene
The lower effective scheme for effectively reducing bandwidth.Especially current Computer Vision speed will in 4K the 60P/S even performance of the above
Ask down, saving Double Data Rate synchronous DRAM DDR bandwidth becomes each Video processing business unit to be needed to consider
One of big factors.
The application scenarios of the technology are main to process (ISP, Image Signal Processing), video in picture signal
Codec (Video codec), video processing subsystem (VPSS, Video Process Sub-System), video show
In image/video processing units such as (VDP, Video display process).By frame buffer compression FBC, per frame pixel count
It is put into DDR according to that can be effectively compressed, subsequent treatment or when using first passes around decompression unit, recovers image pixel data.
Because data have passed through compression treatment, the required data bandwidth of decompression can be saved a lot.First briefly introduce more than each
Demand in scape business.
(1) picture signal treatment ISP is general is being mainly used to the unit to the treatment of front-end image sensor output signal.Than
More typical is exactly the treatment that correlation is carried out in raw image data (RAW DATA) numeric field data.Processing sequence is the light of a frame in
Grid scan (raster scan), and treatment bit-depth is related to chip.Data model storage is packing form (pack mode).
(2) Video codec are mainly used in obtaining reference frame data from DDR, to carry out interframe data prediction.Should
It is complex with scene.The reference frame number for being mainly reflected in each video protocols is relatively more, predicted motion vector (MV) scope
At random, and the MV of two components of brightness and colourity is also differed.So it is accomplished by two and compares crucial demand as follows, with
Machine accesses (random access), internal necessary integrated cache buffer cache resources.It is main to include two cache of type.
Data cache after compression head cache and compression.
(3) video processing subsystem VPSS, video display VDP
Main application scenarios are the grating raster scan process in frame in portion, also there is a small amount of random in VPSS
Access is applied.
Current algorithm is substantially similar, is all based on block to be compressed and decompress.Wherein the block is a pressure
Contracting unit (compression unit).Then, sub-block (sub block), sub are divided into inside compression unit
Block is a real compression treatment unit.The size of compression unit is realized based on algorithm principle, application scenarios, hardware
Each factor such as area, DDR access efficiencies is determined.It is divided into compressed data build compression head and pressure after compression
Two parts of contracting data block compression body, during decoding, complete according to compression head
The decoding of compression body.Compression head mainly comprising each sub-block size, whether compress mark, press
The information such as compressed mode, direct current (DC) value.What compressed data block was mainly stored is the residual error data between pixel.
In the scene of random access, the size head size for data block head are especially sensitive, can directly affect
The size of data block head caching (head cache).Specifically, it is exactly that the number on road (way) of cache can be greatly increased.Such as
Fruit will ensure the performance of more than 4K 60P/S, and cache sizes, hardware realize that sequential (Timing) is all very big bottleneck.
Concrete example is as follows：
For data block head caching head cache, in the case of cache size identicals, improve as far as possible
The hit rate of cache, advantageously ensures that the performance of each processing unit.
For data block body size and DDR burst, it is obvious that the data bits very little of head, about
16--64bit, cache lines cache line are traditionally arranged to be picturedeep for unit.Assuming that algorithm is according to 64x2 block of pixels
Compression (on width 64 pixels, height is 2 rows), in the case of cache miss, the head head that bus can once apply
Maximum image line number be analyzed as follows that (head size are according to most optimistic analysis i.e. 16bits；Picture traverse 4K；Under 4K resolution ratio
The reference frame of coding and decoding video is set to 2).
The data block head (head size) of each 64x2 block of pixels is 16bits, and the compression blocks number of 2 image lines is
4K/64=64, correspondence 64 head, 64x16bits=1024bits are so for burst-length burst length
The bus bit wide of 128bits, it is necessary to burst burst number 1024/128=8, for maximum system AXI burst length
It is to the maximum in the case of 16, cache line are the corresponding head data volumes of 4 image lines.
For 4K images, in the case of there are 2 reference frames according to each frame, a CTB image for 64 sizes is at least buffered
Total lines=(the colourity row of 64 brightness row+32) x2 reference frame numbers (reference that row (64 image lines) needs
Frames)=192 row, the minimum ways=(192lines size/cache line)=48 of cache；Therefore for cache
For, cache way are very big, cause cache size to become big；Way is bigger simultaneously, for hardware realization, sequential timing
It is particularly poor.Timing will be difficult to meet the performance of each processing unit in the case where frequency is improved in the case of very poor
It is required that.
That is the size of head directly governs the resource and hard-wired sequential timing of cache, ultimately results in algorithm
Hard-wired area and performance be all bottleneck.
Current hardware solution mainly has two kinds.
Scheme 1：Compressed data build compression head and compressed data block compression body are separated
Storage, compression body are deposited according to the size of data after compression, bytes registration process.Below with YUV420
The compressed data flow of semi-planar certain component (Y or UV) form storages is as shown in Figure 1.
Each compression unit block, is compressed, each component (Y or UV) after compression by internal compression processing unit
Two regions of DDR, i.e. header field head domain and body domain body domain are placed on, wherein build head is fixed N
Bits, block data body data sizes are not fixed.
During decompression, by first obtaining a content, the body data of correspondence size are then obtained, finally decompressed.
There is following weak point in the program：
Than larger, in general at least 16bits, is substantially in 32bits---64bits the number of bits of build head
Between；As described above, the size of build head directly governs the resource and hard-wired sequential of cache cache
Timing, it is all bottleneck to ultimately result in hard-wired area and performance.
Block data length is not fixed, compression when bus can according to strobe (axi write data protocol in signal, should
Which bytes position that signal designation writes data is effective) it is written in DDR, during decompression, can also be related to strobe and read to ask
Topic.
The advantage of the part is DDR storage space reductions, is also saved with Time Bandwidth.But cannot meet for VPU
Random access random access application.Because data carry strobe alignment problems, the hardware of data cache is realized again
Miscellaneous degree can be greatly increased.The reading of Data Cache will all the time pay close attention to the mark of Strobe.
Scheme 2：
Main pair adjusts with the size (body size) of data block, i.e. the DDR spaces that body size take are
Space in the case of not compressing, such memory size is constant, only reduces bandwidth.
The program greatly reduces hard-wired complexity in the absence of the problem of strobe alignment.But, build head
Size directly govern the resource and hard-wired sequential timing of cache cache, ultimately result in hard-wired
Area and performance are all that bottleneck is still present.
The content of the invention
In view of this, the technical problems to be solved by the invention are to provide one kind and are used for.In order to the embodiment for disclosing
Some aspects have a basic understanding, shown below is simple summary.The summarized section is not extensive overview, nor will
Determine key/critical component or describe the protection domain of these embodiments.Its sole purpose is presented with simple form
Some concepts, in this, as the preamble of following detailed description.
A kind of compression and storage method of video data that the present invention is provided, including：
After video data compression is processed, data block header (compression head) and data block are obtained
(compression body)；
The data block header and data block are stored in predetermined memory space in the lump；
Setting sign is used to indicate corresponding data block compression sizes, and fixed in internal memory according to raster scanning
Bit number is stored；
The storage base unit of compression data block is determined based on bus bandwidth, and is swept according to storage base unit and grating
Retouch order storage compression data block.
Wherein, the data block header and data block are stored in predetermined memory space in the lump, specially by data
Block header and data block and together with, the data block header is using the prefix body prefix of data block as body's
A part.
If the size of the compression data block is KxL, K=2k, k=1, then 2,3 ..., the bit of the sign
Number is：
Wherein K is columns, and L is line number, and N is bit depths of pixels, 2mIt is bus bit wide.
The compression and storage method, also includes：
When carrying out data storage, the specification (body size) of data block after being compressed, from DDR header fields (head
Domain) in obtain the compression after data block specification (the body size of compression unit), further according to the specification
Data block body after compression is obtained, to be placed into corresponding position.
Based on bus broadband, it is determined that the base unit deposited, if bus bandwidth is 2mIndividual bit, then what is deposited is substantially single
Position isIndividual bus bit wide unit, wherein K is columns, and L is line number, and N is bit depths of pixels.
The method further includes the step of searching and extract stored data block：
According to default coordinate offset and vertical shift, after reading the corresponding address data in caching, the number of each address
The sign (HEAD_NEW) of the compression unit can be obtained according to corresponding bits of offset is intercepted；
Specification based on the data block obtains the data block of corresponding length by bus, according to sign
Coordinate, finds corresponding data block.
The present invention also provides a kind of storage device for storing video data, including：
Memory cell for storing video compression data；The video compression data includes data block header and data block
Body；
Memory cell for storing sign (HEAD_NEW), the sign is used to indicate corresponding video compress
The size of data cell, the number of bits of the sign determines according to the size of the compressed data unit；
The video compression data indicates storage according to storage base unit and sign, wherein the storage is basic
Unit is determined based on bus bandwidth.
The compressed data unit deposited in the storage device, its data block header and data block and together with, it is described
Data block header is using the prefix (body prefix) of data block as a part of new data block body.
If the size of the compression data block is KxL, K=2k, k=1, then 2,3 ..., the bit of the sign
Number is：
Wherein K is columns, and L is line number, and N is bit depths of pixels, 2mIt is bus bit wide.
The base unit of the compressed data unit storage determines according to bus broadband, if bus bandwidth is 2mBit
, then the base unit deposited isIndividual highway width unit, wherein K is columns, and L is line number, and N is picture
Plain bit-depth.
In sum, the technical scheme that the present invention is provided, it is pre- by the way that the data block header and data block be stored in
Fixed memory space；Setting sign (HEAD_NEW) is used to indicate effective size of corresponding data block, and gives bus
Bandwidth is alignd.So, under same performance performance requirements, required cache spaces (s ize) are original 1/4, while
Cache roads (way) number is original 1/4, and greatly reducing for road (way) quantity directly results in the reduction of hardware-implemented logic series,
Be conducive to rear end timing closure, frequency higher can be run to, significantly increase performance.
For above-mentioned and related purpose, one or more embodiments include will be explained in below and in claim
In the feature that particularly points out.Following explanation and accompanying drawing describes some illustrative aspects in detail, and its indicate be only
Some modes in the utilizable various modes of principle of each embodiment.Other benefits and novel features will be under
The detailed description in face is considered in conjunction with the accompanying and becomes obvious, the disclosed embodiments be will include all these aspects and they
Be equal to.
Brief description of the drawings
Fig. 1 is a kind of schematic flow sheet for storing compressed data of the prior art；
Fig. 2 is a kind of compression and storage method flow chart of video data provided in an embodiment of the present invention；
Fig. 3 is a kind of schematic flow sheet for storing compressed data provided in an embodiment of the present invention；
Fig. 4 is the schematic diagram of the acquisition reference frame data partial decompressing that the specific embodiment of the invention is provided；
Fig. 5 is the corresponding relation schematic diagram of the address of the image line that the specific embodiment of the invention is provided and cache.
Specific embodiment
The following description and drawings fully show specific embodiments of the present invention, to enable those skilled in the art to
Put into practice them.Other embodiments can include structure, logic, it is electric, process and it is other changes.Embodiment
Only represent possible change.Unless explicitly requested, otherwise single component and function are optional, and the order for operating can be with
Change.The part of some embodiments and feature can be included in or replace part and the feature of other embodiments.This hair
The scope of bright embodiment includes the gamut of claims, and all obtainable of claims is equal to
Thing.Herein, these embodiments of the invention can individually or generally be represented that this is only with term " invention "
For convenience, and if in fact disclosing the invention more than, it is not meant to automatically limit the scope of the application to appoint
What single invention or inventive concept.
In order to realize the purpose of the present invention, the present invention proposes a kind of compression and storage method of video data, reference picture 2,
Comprise the following steps：
S01, after video data compression is processed, obtains data block header (compression head) and data block
(compression body)；
S02, predetermined memory space is stored in by the data block header and data block；
S03, setting sign (HEAD_NEW) is used to indicate the compression sizes of corresponding data block, and in internal memory
Deposited according to raster scan order, fixed number of bits, the number of bits of the sign is according to the big of the compression data block
Small determination；
S04, the storage base unit of compression data block is determined based on bus bandwidth, and according to storage base unit and light
Grid scanning sequency deposits compression data block.
In step S02, the data block header and data block are stored in predetermined memory space, specially by number
According to block header compression head and data block compression body and together with, the data block header
The part of (compression head) using the prefix body prefix of data block as body.
If the size of the compression data block is KxL, K=2k, k=1, then 2,3 ..., the bit of the sign
Number is：
Wherein K is columns, and L is line number, and N is bit depths of pixels, 2mIt is bus bit wide.
The compression and storage method, specifically includes：
When carrying out data storage, the specification (body size) of data block after being compressed, from DDR head domain
The specification (the body size of compression unit) of data block after the compression is obtained in domain, is obtained further according to the specification
Data block body after compression, to be placed into corresponding position.
Based on bus broadband, it is determined that the base unit deposited, if bus bandwidth is 2mBit, the then base unit deposited
ForIndividual byte, wherein 2kIt is columns, L is line number, and N is bit depths of pixels, and [] represents and round.
Further, the method also includes the step of searching and extract stored data block：
According to default coordinate offset and vertical shift, after reading the corresponding address data in caching, the number of each address
The sign (HEAD_NEW) of the compression unit can be obtained according to corresponding bits of offset is intercepted；
Specification (body size) based on the data block obtains the data block of corresponding length, root by bus
According to the coordinate of sign, corresponding data block is found.
The present invention also provides a kind of storage device for storing video data, including：
Memory cell for storing video compression data；The video compression data includes data block header
(compression head) and data block (compression body)；
Memory cell for storing sign (HEAD_NEW), the sign is used to indicate corresponding video compress
The compression sizes of data cell, and deposited according to raster scan order fixed number of bits in internal memory, the sign
Number of bits determines according to the size of the compressed data unit.
The data block header (compression head) and data block (compression body) and together,
The data block header (compression head) is using the prefix body prefix of data block as new data block body
A part.
The video compression data indicates storage according to storage base unit and sign, wherein the storage is basic
Unit is determined based on bus bandwidth.The size offset space of data block body is according to original compression between each compression unit
Opened up under unit (not compressing situation), and be also to be stored according to raster scan order.
If the size of the compression data block is KxL, K=2k, then the number of bits of the sign be：
Wherein K is columns, and L is line number, and N is bit depths of pixels, 2mIt is bus bit wide.
The base unit of the compressed data unit storage determines according to bus broadband, if bus bandwidth is 2iBit
, then the base unit deposited isIndividual byte, wherein K are columns, and L is line number, and N is pixel bit depth
Degree.
In order that principle of the invention, characteristic and advantage are clearer, with reference to specific implementation to of the invention
Technical scheme is described in detail.
Reference picture 3, the compressed data build compression head that common compression algorithm unit is exported are incorporated into
In compressed data body body, i.e., using block prefix body prefix as a part of body, body originally is used as body
core.After being incorporated to, because bus access is at least 16bytes alignment, the bandwidth increase of body almost can be ignored.So
, it is necessary to sign (HEAD_NEW) information being known a priori by is little for hardware realization, as long as block after being compressed
The size of body.Block after the compression of the compression unit is so first obtained from DDR header field head domain
Specification body size, then further according to the block specification obtain data block, finally decode.And for the specification,
If bus broadband is 128bits, then be exactly 16bytes alignment.Base unit is 16bytes, rather than with bytes pairs
Together.Therefore the content of the sign (HEAD_NEW) for setting is as follows：
For YUV420semi-planar, because brightness and colourity are stored separately separate compression.Compress single with 64x2
As a example by position, alignd using 16bytes.It is maximum because 64x2 maximums in the case of uncompressed are 64x2byts=8x16byts
It is 8 base units, 3bit can be represented.
Realized for the ease of hardware, a reserved bit reserve bit, i.e., newly-installed sign can be extended
HEAD_NEW is 4 bits.Corresponding, according to analysis before, for 4K frames, the cache line are 16 image lines (i.e. 8
Head rows, because the height of each compression unit is 2 rows) HEAD_NEW data volumes.
In a particular application, as in video decoding Video codec, prediction branch obtains reference frame data partial decompressing
Implementation, reference picture 4 is analyzed as follows：
(1) the reference prediction data block (also referred to as prediction data unit PU) that current decoding needs is assumed, it is assumed that reference block
Size is 16x4, and the row (cur_line) of the reference block of current request is the 0th row, and its coordinate is P (x, y)=P (A, B), then
It is U (x, y)=U ((A that the position that the corresponding compression unit block 64x2 of the row is unit, i.e. coordinate can be found>>2),(B>>
1))；
(2) judge whether cache hits according to U (x, y) coordinate；Assuming that cache is in the case of miss, obtain first
HEAD_NEW data are written in HEAD_NEW cache, and now according to coordinate, and the size of HEAD_NEW is fixed
4bit, can obtain the HEAD_NEW data (16 image line=8 head rows) of correspondence cache line, as shown in Figure 4
HEAD_NEW rows 0 to row 7；The corresponding Cache line of current line of current predictive data cell (PU).
(3) and then according to current line (cur_line) coordinate the corresponding HEAD_NEW in cache is read；
Wherein (2) (3), corresponding relation such as Fig. 5 institutes of the address of image line (HEAD_NEW line in other words) and cache
Show.
Write cache processes：In upper figure, the bits numbers that one head line of 4K images takes are：4K/64*4bit=
256bit=2x128bits；I.e. 2 bus burst；2 cache addresses of i.e. one head line correspondence；I.e. figure 1 above/2 are marked
Note；So whole 16 bus burst correspondences addr0---addr15.
Read cache processes：Each address takes 1/2head line, while levels of the U (x, y) in each head line
Coordinate offset and vertical shift are also known, therefore can be by after the corresponding address data for reading cache, the number of each address
The HEAD_NEW of the compression unit is obtained with according to the corresponding bits of offset of interception.
(4) after HEAD_NEW is obtained, the data block of corresponding length is obtained by bus according to data block specifications size
Body, it is because the internal memory shared by data block body is identical with not compressing the internal memory that occupies, i.e. shared after each 64x2 compressions
Some spaces or 64x2bytes, therefore according to the coordinate of sign HEAD_NEW, corresponding data block can be found
body.So, it will be apparent that save bandwidth.
(5) it is last, start decoding compression body prefix (body_prefix) after acquisition body, and then according to body_
Prefix contents and compression body core (body_core) content, recover image pixel data.
For above-mentioned each scheme, contrast as follows by taking 64x2 compression units as an example, under performance requirement same case：
| Prior art | The present invention | |
| HEAD_NEW bits number of bits | At least 16bits | 4bits |
| Required Cache spaces size | At least 4N | N |
| Required Cache roads (way) number | More than 48, representative value 64 | More than 12, representative value 16 |
Conclusion：The technical scheme that the present invention is provided, under same performance performance requirements, required cache spaces
Size is original 1/4, while cache roads (way) number is original 1/4, greatly reducing for road way quantity directly results in hardware
Realize that logic series is reduced, be conducive to rear end timing closure, frequency higher can be run to, significantly increase performance.
Those skilled in the art are it will be appreciated that here in conjunction with the various exemplary method step of the disclosed embodiments description
Rapid and device unit can be realized with electronic hardware, software or the combination of the two.In order to be clearly shown that hardware and software it
Between interchangeability, the description that the above carries out generally to various exemplary steps and unit in its functional form.
This feature is specifically to apply the design realized with whole system about with hardware realization or implemented in software depending on
Beam.Those skilled in the art can be directed to each specific application, realize described feature in many ways, but this
The result realized is planted to should not be construed as away from the scope of the present invention.
Using general processor, digital signal processor (DSP), application specific integrated circuit (ASIC), field-programmable gate array
Row (FPGA) or other programmable logical device, discrete gate or transistor logics, discrete hardware components or they it
In any combination, it is possible to achieve or perform combine embodiment disclosed herein description various exemplary unit.General place
Reason device be probably microprocessor, but in another scenario, the processor be probably any conventional processor, controller,
Microcontroller or state machine.Processor is likely to be implemented as the combination of computing device, for example, the group of DSP and microprocessor
Conjunction, multi-microprocessor, the microprocessor of one or more combination DSP cores or any other this kind of structure.
The step of method with reference to described by embodiment disclosed above can be embodied directly in hardware, by computing device
Software module or the combination of both.Software module is likely to be present in RAM memory, flash memory, ROM memory, EPROM storages
The storage of device, eeprom memory, register, hard disk, mobile disk, CD-ROM or any other form well known in the art
In medium.A kind of exemplary storage medium is coupled with processor, so that processor can read information from the storage media, and
Can be to the storage media write information.In example is replaced, storage media is the part of processor.Processor and storage media
It is likely to be present in an ASIC.The ASIC is likely to be present in a subscriber station.In replacing example at one, processor and deposit
Storage medium can exist as the discrete assembly in subscriber station.
According to the disclosed embodiment, those skilled in the art can be enabled to realize or use the present invention.It is right
For those skilled in the art, the various modifications of these embodiments are it will be apparent that and the general principles that define here
Other embodiment can also be applied on the basis of the scope and spirit of the present invention are not departed from.Embodiment described above is only
Presently preferred embodiments of the present invention, is not intended to limit the invention, it is all within the spirit and principles in the present invention, made appoint
What modification, equivalent, improvement etc., should be included within the scope of the present invention.
Claims (10)
1. a kind of compression and storage method of video data, it is characterised in that including：
After video data compression is processed, data block header (compression head) and data block are obtained
(compression body)；
The data block header and data block are stored in predetermined memory space in the lump；
Setting sign is used to indicate corresponding data block compression sizes, and according to raster scanning in internal memory fixed bit
Number storage；
The storage base unit of compression data block is determined based on bus bandwidth, and it is suitable according to storage base unit and raster scanning
Sequence deposits compression data block.
2. compression and storage method as claimed in claim 1, it is characterised in that described by the data block header and data block
Be stored in predetermined memory space in the lump, specially by data block header and data block and together with, the data block header
Using the prefix body prefix of data block as a part of data block (body).
3. compression and storage method as claimed in claim 1, it is characterised in that if the size of the compression data block is KxL, K
=2k, k=1,2,3 ..., then the number of bits of the sign be：
Wherein K is columns, and L is line number, and N is bit depths of pixels, 2mIt is bus bit wide.
4. compression and storage method as claimed in claim 1, it is characterised in that also include：
When carrying out data storage, the specification (body size) of data block after being compressed, from DDR header fields (head domain)
The specification (the body size of compression unit) of data block, obtains further according to the specification and presses after middle acquisition compression
Data block body after contracting, to be placed into corresponding position.
5. compression and storage method as claimed in claim 4, it is characterised in that
Based on bus broadband, it is determined that the base unit deposited, if bus bandwidth is 2mIndividual bit, then the base unit deposited isIndividual bus bit wide unit, wherein K is columns, and L is line number, and N is bit depths of pixels.
6. compression and storage method as claimed in claim 4, it is characterised in that further include to search and extract stored number
The step of according to block：
According to default coordinate offset and vertical shift, after reading the corresponding address data in caching, the data of each address are cut
The sign (HEAD_NEW) of the compression unit is obtained by taking corresponding bits of offset；
Specification based on the data block obtains the data block of corresponding length by bus, according to the seat of sign
Mark, finds corresponding data block.
7. a kind of storage device for storing video data, it is characterised in that including：
Memory cell for storing video compression data；The video compression data includes data block header and data block；
Memory cell for storing sign (HEAD_NEW), the sign is used to indicate corresponding video compression data
The size of unit, the number of bits of the sign determines according to the size of the compressed data unit；
The video compression data indicates storage according to storage base unit and sign, wherein the storage base unit
It is to be determined based on bus bandwidth.
8. storage device as claimed in claim 7, it is characterised in that
The data block header and data block and together with, the data block header is with the prefix (body of data block
Prefix) as a part of new data block body.
9. storage device as claimed in claim 7, it is characterised in that if the size of the compression data block is KxL, K=2k, k
=1,2,3 ..., then the number of bits of the sign be：
Wherein K is columns, and L is line number, and N is bit depths of pixels, 2mIt is bus bit wide.
10. storage device as claimed in claim 7, it is characterised in that the base unit root of the compressed data unit storage
Determine according to bus broadband, if bus bandwidth is 2mBit, then the base unit deposited isIndividual bus
Width unit, wherein K are columns, and L is line number, and N is bit depths of pixels.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201611237039.0A CN106713926B (en) | 2016-12-28 | 2016-12-28 | Compression storage method and device for video data |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201611237039.0A CN106713926B (en) | 2016-12-28 | 2016-12-28 | Compression storage method and device for video data |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN106713926A true CN106713926A (en) | 2017-05-24 |
| CN106713926B CN106713926B (en) | 2021-08-03 |
Family
ID=58903639
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201611237039.0A Active CN106713926B (en) | 2016-12-28 | 2016-12-28 | Compression storage method and device for video data |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106713926B (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113330748A (en) * | 2019-02-07 | 2021-08-31 | 华为技术有限公司 | Method and apparatus for intra prediction mode signaling |
| CN115550669A (en) * | 2022-11-30 | 2022-12-30 | 摩尔线程智能科技(北京)有限责任公司 | Video transcoding method and device, electronic equipment and storage medium |
| CN117708071A (en) * | 2024-02-06 | 2024-03-15 | 苔花科迈(西安)信息技术有限公司 | Processing method and device for coal mine equipment operation parameters based on big data |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102053880A (en) * | 2011-01-07 | 2011-05-11 | 杭州电子科技大学 | Rar file carving recovery method based on contents |
| CN102096698A (en) * | 2010-12-14 | 2011-06-15 | 青岛海信网络科技股份有限公司 | Video data storage format, storage method and search method |
| JP2012124504A (en) * | 2005-02-02 | 2012-06-28 | Sony Chemical & Information Device Corp | Method of implementing electrical component |
| CN103366006A (en) * | 2013-07-19 | 2013-10-23 | 武汉睿数信息技术有限公司 | Method for establishing geography space index based on three dimensional laser-point cloud |
-
2016
- 2016-12-28 CN CN201611237039.0A patent/CN106713926B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2012124504A (en) * | 2005-02-02 | 2012-06-28 | Sony Chemical & Information Device Corp | Method of implementing electrical component |
| CN102096698A (en) * | 2010-12-14 | 2011-06-15 | 青岛海信网络科技股份有限公司 | Video data storage format, storage method and search method |
| CN102053880A (en) * | 2011-01-07 | 2011-05-11 | 杭州电子科技大学 | Rar file carving recovery method based on contents |
| CN103366006A (en) * | 2013-07-19 | 2013-10-23 | 武汉睿数信息技术有限公司 | Method for establishing geography space index based on three dimensional laser-point cloud |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113330748A (en) * | 2019-02-07 | 2021-08-31 | 华为技术有限公司 | Method and apparatus for intra prediction mode signaling |
| CN113330748B (en) * | 2019-02-07 | 2023-04-18 | 华为技术有限公司 | Method and apparatus for intra prediction mode signaling |
| CN115550669A (en) * | 2022-11-30 | 2022-12-30 | 摩尔线程智能科技(北京)有限责任公司 | Video transcoding method and device, electronic equipment and storage medium |
| CN117708071A (en) * | 2024-02-06 | 2024-03-15 | 苔花科迈(西安)信息技术有限公司 | Processing method and device for coal mine equipment operation parameters based on big data |
| CN117708071B (en) * | 2024-02-06 | 2024-05-28 | 苔花科迈(西安)信息技术有限公司 | Processing method and device for coal mine equipment operation parameters based on big data |
Also Published As
| Publication number | Publication date |
|---|---|
| CN106713926B (en) | 2021-08-03 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US7039241B1 (en) | Method and apparatus for compression and decompression of color data | |
| CN105933708B (en) | A kind of method and apparatus of data compression and decompression | |
| US11023152B2 (en) | Methods and apparatus for storing data in memory in data processing systems | |
| JP5969914B2 (en) | Video compression / decompression device | |
| CN106030652B (en) | Method, system and composite display controller for providing output surface and computer medium | |
| CN106713926A (en) | Compression storage method and device for video data | |
| US5862264A (en) | Image compression device and image compression method for compressing images by smoothing them, reversibly compressing the residual images and compressing the smoothed images based on fractal theory | |
| CN105578190A (en) | Lossless compression method and system for video hard decoding | |
| CN110728725B (en) | Hardware-friendly real-time system-oriented lossless texture compression method | |
| CN114501024B (en) | Video compression system, method, computer readable storage medium and server | |
| CN104952088A (en) | Method for compressing and decompressing display data | |
| CN105472389A (en) | Out-chip buffer compression method for superhigh-definition processing system | |
| JP2009211494A (en) | Information processor, and information processing method | |
| US10824357B2 (en) | Updating data stored in a memory | |
| CN106651971A (en) | Graphic data compression device and graphic data compression method | |
| CN102263946B (en) | Techniques for storing and retrieving pixel data | |
| US20120147023A1 (en) | Caching apparatus and method for video motion estimation and compensation | |
| CN110322904A (en) | Compress image information reading and control method thereof and device | |
| US10418002B2 (en) | Merged access units in frame buffer compression | |
| JPH04326669A (en) | Picture encoding device | |
| US11205243B2 (en) | Data processing systems | |
| CN108876703B (en) | Data storage method | |
| KR100753504B1 (en) | In-loop Deblocking Filter System and method for filterring data using the same | |
| CN110392259A (en) | Image compression system and utilization image compression system are in the method for compressing image picture | |
| CN101257590B (en) | Coding, decoding method and device, display system |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| CB02 | Change of applicant information |
Address after: Room E203, Building No. 76, West Fourth Ring South Road, Fengtai District, Beijing 100071 Applicant after: New Shoreline (Beijing) Science and Technology Group Co.,Ltd. Address before: Room E203, Building No. 76, West Fourth Ring South Road, Fengtai District, Beijing 100071 Applicant before: Beijing New Shoreline Technology Co.,Ltd. Address after: Room E203, Building No. 76, West Fourth Ring South Road, Fengtai District, Beijing 100071 Applicant after: Beijing New Shoreline Technology Co.,Ltd. Address before: Room E203, Building No. 76, West Fourth Ring South Road, Fengtai District, Beijing 100071 Applicant before: BEIJING PUJIXIN TECHNOLOGY CO.,LTD. |
|
| CB02 | Change of applicant information | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |